Re: [PATCH v2] gpu: GPU_BUDDY_KUNIT_TEST should not select GPU_BUDDY

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Adding `depends on DRM || COMPILE_TEST` to the `bool` symbol `GPU_BUDDY` causes a Kconfig unmet direct dependencies warning when `DRM=m`, and breaks manual selection of the KUnit test.
--

--- Patch [1]: [PATCH v2] gpu: GPU_BUDDY_KUNIT_TEST should not select GPU_BUDDY ---
commit 0c0f9dc26f6481afe68d4de6c66198acb6cb9883
Author: Geert Uytterhoeven <geert@linux-m68k.org>

gpu: GPU_BUDDY_KUNIT_TEST should not select GPU_BUDDY

This commit changes GPU_BUDDY_KUNIT_TEST to depend on GPU_BUDDY instead
of selecting it. It also adds a dependency on DRM or COMPILE_TEST to
GPU_BUDDY and makes it user-visible when KUNIT is enabled.

> diff --git a/drivers/gpu/Kconfig b/drivers/gpu/Kconfig
> index 3bcf08260d73a..d881ab6f7a006 100644
> --- a/drivers/gpu/Kconfig
> +++ b/drivers/gpu/Kconfig
> @@ -1,14 +1,15 @@
>  # SPDX-License-Identifier: GPL-2.0
>  
>  config GPU_BUDDY
> -       bool
> +       bool "Page based buddy allocator for GPU memory" if KUNIT
> +       depends on DRM || COMPILE_TEST

[Severity: Low]
Does adding this dependency cause a Kconfig warning when DRM is configured
as a module?

Because GPU_BUDDY is a boolean symbol, this dependency restriction evaluates
to 'm' when CONFIG_DRM=m and CONFIG_COMPILE_TEST=n. This causes a conflict
with the 'select' from DRM_BUDDY, resulting in an unmet direct dependencies
warning.

Additionally, could this prevent developers from manually enabling the KUnit
test when DRM=m without COMPILE_TEST?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/e67a2cb0d599b75ae365882913c543e7b0c5b8a3.1787671708.git.geert@linux-m68k.org?part=1